Inertial sensors are used in a wide range of applications for tracking the movement of objects, such as limbs, cameras, input devices, or head mounted displays (HMDs). In some examples, inertial tracking devices have been successfully applied to a wide range of applications including virtual environment (VE) training, virtual prototyping, interactive visualization and design, virtual reality (VR) gaming, and vehicle simulation. Despite the level of accuracy provided by some inertial sensors, some existing technologies cannot be used in certain applications. For example, when an inertial sensor is used to track the movement of an object in a moving vehicle, the signal produced by the inertial sensor cannot be relied upon because the inertial sensor cannot determine if inertial forces are caused by the vehicle or the object.
To overcome some of the shortcomings of existing technologies, some devices configured with an inertial sensor are enhanced with visual tracking sensors. For example, in addition to using an inertial sensor, some devices have one or more cameras to track the movement of an object. However, solutions using visual tracking sensors are not as efficient as solutions using inertial sensors when it comes to computing resources and power consumption.
It is with respect to these and other considerations that the disclosure made herein is presented.